Uncovering Latent Bias in LLM-Based Emergency Department Triage Through Proxy Variables

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autor principal:	Zhang, Ethan
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

From Promising Capability to Pervasive Bias: Assessing Large Language Models for Emergency Department Triage
por: Lee, Joseph, et al.
Publicado: (2025)

Support Systems of Clinical Decisions in the Triage of the Emergency Department Using Artificial Intelligence: The Efficiency to Support Triage
por: Eleni Karlafti
Publicado: (2023)

How Does AI Improve Emergency Department Triage?
por: Rasit Dinc
Publicado: (2017)

Development of a Large Language Model-based Multi-Agent Clinical Decision Support System for Korean Triage and Acuity Scale (KTAS)-Based Triage and Treatment Planning in Emergency Departments
por: Han, Seungjun, et al.
Publicado: (2024)

Validating Emergency Department Admission Predictions Based on Local Data Through MIMIC-IV
por: Meimeti, Francesca, et al.
Publicado: (2025)

VulTriage: Triple-Path Context Augmentation for LLM-Based Vulnerability Detection
por: Tang, Wenxin, et al.
Publicado: (2026)

Enhancing Vaccine Safety Surveillance: Extracting Vaccine Mentions from Emergency Department Triage Notes Using Fine-Tuned Large Language Models
por: Khademi, Sedigh, et al.
Publicado: (2025)

T2IBias: Uncovering Societal Bias Encoded in the Latent Space of Text-to-Image Generative Models
por: Sufian, Abu, et al.
Publicado: (2025)

EQUITRIAGE: A Fairness Audit of Gender Bias in LLM-Based Emergency Department Triage
por: Young, Richard J., et al.
Publicado: (2026)

BiasBusters: Uncovering and Mitigating Tool Selection Bias in Large Language Models
por: Blankenstein, Thierry, et al.
Publicado: (2025)

An Adaptive Simulated Annealing-Based Machine Learning Approach for Developing an E-Triage Tool for Hospital Emergency Operations
por: Ahmed, Abdulaziz, et al.
Publicado: (2022)

Uncovering Latent Chain of Thought Vectors in Language Models
por: Zhang, Jason, et al.
Publicado: (2024)

Implicit Geographic Inference in LLM Medical Triage: Language-Driven Disparities in Emergency Recommendations
por: Wong, Qi Han
Publicado: (2026)

Uncovering Latent Communication Patterns in Brain Networks via Adaptive Flow Routing
por: Huang, Tianhao, et al.
Publicado: (2026)

Fast Proxies for LLM Robustness Evaluation
por: Beyer, Tim, et al.
Publicado: (2025)

Steering LLM Reasoning Through Bias-Only Adaptation
por: Sinii, Viacheslav, et al.
Publicado: (2025)

Uncovering the Latent Potential of Deep Intermediate Representations
por: Batra, Arnesh, et al.
Publicado: (2026)

Fast and Accurate Explanations of Distance-Based Classifiers by Uncovering Latent Explanatory Structures
por: Bley, Florian, et al.
Publicado: (2025)

ProxyKV: Cross-Model Proxy Pruning for Efficient Long-Context LLM Inference
por: Li, Junjie, et al.
Publicado: (2026)

Social Evolution of Published Text and The Emergence of Artificial Intelligence Through Large Language Models and The Problem of Toxicity and Bias
por: Khan, Arifa, et al.
Publicado: (2024)

Syn-STARTS: Synthesized START Triage Scenario Generation Framework for Scalable LLM Evaluation
por: Hagiwara, Chiharu, et al.
Publicado: (2025)

Synthesizing Images on Perceptual Boundaries of ANNs for Uncovering and Manipulating Human Perceptual Variability
por: Wei, Chen, et al.
Publicado: (2025)

Uncovering the Spectral Bias in Diagonal State Space Models
por: Solozabal, Ruben, et al.
Publicado: (2025)

Long-Context Reasoning Through Proxy-Based Chain-of-Thought Tuning
por: Li, Miao, et al.
Publicado: (2026)

A Framework for Analyzing Abnormal Emergence in Service Ecosystems Through LLM-based Agent Intention Mining
por: Shen, Yifan, et al.
Publicado: (2025)

Uncovering Political Bias in Large Language Models using Parliamentary Voting Records
por: Chen, Jieying, et al.
Publicado: (2026)

Predicting Elevated Risk of Hospitalization Following Emergency Department Discharges
por: Hong, Dat, et al.
Publicado: (2024)

Policy-Guided Threat Hunting: An LLM enabled Framework with Splunk SOC Triage
por: Sahay, Rishikesh, et al.
Publicado: (2026)

Internal Representation, Not Clinical Knowledge: Where Apparent LLM Triage Failures Originate
por: Navarro, David Fraile, et al.
Publicado: (2026)

Toward Scalable Verifiable Reward: Proxy State-Based Evaluation for Multi-turn Tool-Calling LLM Agents
por: Chuang, Yun-Shiuan, et al.
Publicado: (2026)

Uncovering Bias in Foundation Models: Impact, Testing, Harm, and Mitigation
por: Sun, Shuzhou, et al.
Publicado: (2025)

Predicting LLM Reasoning Performance with Small Proxy Model
por: Koh, Woosung, et al.
Publicado: (2025)

UniCog: Uncovering Cognitive Abilities of LLMs through Latent Mind Space Analysis
por: Liu, Jiayu, et al.
Publicado: (2026)

Model Consistency as a Cheap yet Predictive Proxy for LLM Elo Scores
por: Ramaswamy, Ashwin, et al.
Publicado: (2025)

State-Augmented Graphs for Circular Economy Triage
por: Fox, Richard, et al.
Publicado: (2025)

Uncovering Bias in Large Vision-Language Models with Counterfactuals
por: Howard, Phillip, et al.
Publicado: (2024)

Uncovering Implicit Bias in Large Language Models with Concept Learning Dataset
por: Wang, Leroy Z.
Publicado: (2025)

Don't Start What You Can't Finish: A Counterfactual Audit of Support-State Triage in LLM Agents
por: Unlu, Eren
Publicado: (2026)

TriAgent: Automated Biomarker Discovery with Deep Research Grounding for Triage in Acute Care by LLM-Based Multi-Agent Collaboration
por: Delikoyun, Kerem, et al.
Publicado: (2025)

Small Language Models for Emergency Departments Decision Support: A Benchmark Study
por: Wang, Zirui, et al.
Publicado: (2025)

if (filemtime($filename) + $maxlifetime < time()) { unlink($filename); $count++; } } return $count; } /** * A function that is called internally when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ protected function saveSession($sessId, $data): bool { $sessFile = $this->path . '/sess_' . $sessId; if ($handle = fopen($sessFile, 'w')) { $return = false; // Lock the file for exclusive access to avoid issues with multiple // processes writing session simultaneously: if (flock($handle, LOCK_EX)) { $return = fwrite($handle, $data); // Make sure that there's no trailing data by truncating the file to // the correct length: ftruncate($handle, strlen($data)); fflush($handle); flock($handle, LOCK_UN); } fclose($handle); if ($return !== false) { return true; } } // If we got this far, something went wrong with the file output... // It is tempting to throw an exception here, but this code is called // outside of the context of exception handling, so all we can do is // echo a message.

// Anecdotal testing Today and Yesterday seems to indicate destroy() // is called by the garbage collector and everything is good. // Something to keep in mind though. return 0; } /** * Write function that is called when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ public function write($sessId, $data): bool { if ($this->writesDisabled) { return true; } return $this->saveSession($sessId, $data); } /** * A function that is called internally when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ abstract protected function saveSession($sessId, $data): bool; }

* * @return void */ public function writeClose() { // The assumption is that we're using PHP's ext/session. // session_write_close() will actually overwrite $_SESSION with an // empty array on completion -- which leads to a mismatch between what // is in the storage object and $_SESSION. To get around this, we // temporarily reset $_SESSION to an array, and then re-link it to // the storage object. // // Additionally, while you _can_ write to $_SESSION following a // session_write_close() operation, no changes made to it will be // flushed to the session handler. As such, we now mark the storage // object isImmutable. $storage = $this->getStorage(); if (! $storage->isImmutable()) { $_SESSION = $storage->toArray(true); session_write_close(); $storage->fromArray($_SESSION); $storage->markImmutable(); } } /** * Attempt to set the session name * * If the session has already been started, or if the name provided fails * validation, an exception will be raised. * * @param string $name * @return SessionManager * @throws Exception\InvalidArgumentException */ public function setName($name) { if ($this->sessionExists()) { throw new Exception\InvalidArgumentException( 'Cannot set session name after a session has already started'

Environment & details:

GET Data empty

POST Data empty

Files empty

Key	Value
language	"es"
VUFIND_SESSION	"8db470839888287d2de8f50ab0d9ae28"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790839263.3455 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "8db470839888287d2de8f50ab0d9ae28" ]`]
SessionState	Laminas\Stdlib\ArrayObject {#938}
FlashMessenger	Laminas\Stdlib\ArrayObject {#1433}
SessionHelper	Laminas\Stdlib\ArrayObject {#1759}

Server/Request Data

Key	Value
REDIRECT_VUFIND_LOCAL_DIR	"/usr/local/vufind/local"
REDIRECT_HTTPS	"on"
REDIRECT_SSL_TLS_SNI	"info.umar.mx"
REDIRECT_STATUS	"200"
VUFIND_LOCAL_DIR	"/usr/local/vufind/local"
HTTPS	"on"
SSL_TLS_SNI	"info.umar.mx"
HTTP_ACCEPT	"/"
HTTP_USER_AGENT	"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)"
HTTP_COOKIE	"language=es; VUFIND_SESSION=8db470839888287d2de8f50ab0d9ae28; ui=standard"
HTTP_ACCEPT_ENCODING	"gzip, br, zstd, deflate"
HTTP_HOST	"info.umar.mx"
PATH	"/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/snap/bin"
SERVER_SIGNATURE	"<address>Apache/2.4.66 (Ubuntu) Server at info.umar.mx Port 443</address>\n"
SERVER_SOFTWARE	"Apache/2.4.66 (Ubuntu)"
SERVER_NAME	"info.umar.mx"
SERVER_ADDR	"200.23.222.19"
SERVER_PORT	"443"
REMOTE_ADDR	"216.73.216.147"
DOCUMENT_ROOT	"/usr/local/vufind/public"
REQUEST_SCHEME	"https"
CONTEXT_PREFIX	""
CONTEXT_DOCUMENT_ROOT	"/usr/local/vufind/public"
SERVER_ADMIN	"admin@localhost"
SCRIPT_FILENAME	"/usr/local/vufind/public/index.php"
REMOTE_PORT	"61644"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2601_15306/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2601_15306/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790839263.3414
REQUEST_TIME	1790839263

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler